implementation of a tms-fmri system: a primer · 2021. 5. 19. · navarro de lara et al., 2015;...

1

Implementation of a TMS-fMRI system: A primer

Golnoush Alamian, PhD1,2†, Ruiyang Ge, PhD1,2†; Afifa Humaira, BS1,2; Elizabeth

Gregory, MSc1,2, Erin L. MacMillan PhD3,4, Laura Barlow RTR, RTMR5, Fidel Vila-

Rodriguez, MD, PhD1,2

1Non-Invasive Neurostimulation Therapies (NINET) Laboratory, Department of

Psychiatry, University of British Columbia, Vancouver, BC, Canada 2Djavad Mowafaghian Centre for Brain Health, University of British Columbia,

Vancouver, BC, Canada 3School of Mechatronic Systems Engineering, Faculty of Applied Sciences, Simon

Fraser University, Vancouver, British Columbia, Canada; 4Department of Radiology, University of British Columbia, Vancouver, British Columbia,

Canada 5UBC MRI Research Centre, University of British Columbia, Vancouver, BC, Canada

* Correspondence to: Fidel Vila-Rodriguez, MD, PhD, FRCP, FAPA

Non-Invasive Neurostimulation Therapies (NINET) Laboratory

Department of Psychiatry, University of British Columbia

2255 Wesbrook Mall, Vancouver, BC V6T 2A1, Canada

Tel: 604-822-7512

E-mail: [email protected]

.CC-BY-NC-ND 4.0 International licenseavailable under awas not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made

The copyright holder for this preprint (whichthis version posted May 21, 2021. ; https://doi.org/10.1101/2021.05.19.444738doi: bioRxiv preprint

https://doi.org/10.1101/2021.05.19.444738

http://creativecommons.org/licenses/by-nc-nd/4.0/

2

ABSTRACT

Transcranial magnetic stimulation (TMS) is a non-invasive and non-pharmacological

intervention, approved for the treatment of individuals diagnosed with treatment-

resistant depression. This well-tolerated approach uses magnetic pulses to stimulate

specific brain regions and induce changes in brain networks at multiple levels of human

functioning. Combining TMS with other neuroimaging techniques, such as functional

magnetic resonance imaging (fMRI), offers new insights into brain functioning, and

allows to map out the causal alterations brought on by TMS interventions on neural

network connectivity and behaviour. However, the implemention of concurrent TMS-

fMRI brings on a number of technical challenges that must be overcome to ensure good

quality of functional images. The goal of this study was thus to investigate the impact of

TMS pulses in an MR-environment on the quality of BRAINO phantom images, in terms

of the signal of the images, the temporal fluctuation noise, the spatial noise and the

signal to fluctuation noise ratio, at the University of British Columbia (UBC)

Neuroimaging facility. The results of our analyses replicated those of previous sites, and

showed that the present set-up for concurrent TMS-fMRI ensures minimal noise artefact

on functional images obtained through this multimodal approach. This step was a key

stepping stone for future clinical trials at UBC.



https://doi.org/10.1101/2021.05.19.444738


3

1. Introduction

1.1. Background

Psychiatric diseases such as major depressive disorder (MDD) are typically treated

with a combination of psychotherapy and pharmaceutical treatments. This first response

route however fails to improve the symptoms in about half of these patients. After two

rounds of failed drug-trial treatment, affected individuals are categorized as having

treatment resistant depression (TRD) (Conway et al., 2017). It is estimated that more

than a third of MDD patients fail to achieve remission using antidepressants and are

thus classified as having TRD (Nemeroff, 2007).

Over the years, this account has given way to alternative treatment options.

Transcranial magnetic stimulation (TMS) (George et al., 1999; Hallett, 2000; Vila-

Rodriguez et al., 2013), for instance, is a non-invasive and well-tolerated intervention,

which has received both Health Canada and US FDA approval for the treatment of

TRD. This approach relies on electromagnetic induction to stimulate brain regions

focally, and induce changes in brain function at the cognitive, motor, emotional and

behaviour levels. Specifically, TMS changes the brain’s electric current in target brain

areas by sending magnetic pulses using a coil, which change the local field potentials at

the surface of the cortex, thereby changing the excitatory/inhibitory neuronal activity

within a given region (Downar et al., 2016). Of note, given its nature, TMS allows to

investigate causality in the brain, by measuring functional changes associated with the

electromagnetic stimulation, something that is inaccessible through other human brain

mapping methods (Siebner et al., 2009). Hence, combining TMS interventions with a

neuroimaging technique could reveal the critical temporal and spatial neural dynamics

involved in the causal network changes that occur during TMS treatment in neurological

and psychiatric patient populations.



https://doi.org/10.1101/2021.05.19.444738


4

One neuroimaging method that has long-been used for the study of brain activity and

connectivity patterns is functional magnetic resonance imaging (fMRI). fMRI is a

powerful tool that measures changes in the brain’s concentration of oxygenated and

deoxygenated hemoglobin via the blood-oxygen-level-dependent (BOLD) signal

(Huettel et al., 2004; Buxton, 2009). It is thought that the BOLD signal changes as a

result of neuronal activity given that is coupled with cerebral blood flow and energy use

(Logothetis, 2008). Moreover, given its’ precise millimetre spatial resolution and fair

temporal resolution, fMRI is the most widely used neuroimaging technique to investigate

brain function.

Concurrent TMS-fMRI is an emerging imaging approach that combines these two

methodologies, and allows for TMS treatment inside the MR-scanner, while acquiring

fMRI data. By employing this technique, cortical excitability can be manipulated, while

monitoring the intra- and inter-cerebral functional activity of a brain region.

1.1.1. Summary of previous TMS-fMRI experiments in healthy and clinical

populations

Concurrent TMS-fMRI was first performed by (Bohning et al., 1997), and

experiments have been on the rise in both healthy (Bohning et al., 1997; Nahas et al.,

2001; Denslow et al., 2005; Bestmann et al., 2008b; de Vries et al., 2009; Moisa et al.,

2009, 2010; Li et al., 2011; Ricci et al., 2012; Hanlon et al., 2013; Schintu et al., 2021;

Oathes et al., 2021) and clinical populations (Li et al., 2004; Bestmann et al., 2006; de

Vries et al., 2012; Webler et al., 2020). This is not surprising given that this multimodal

approach allows to, not only examine the altered brain connectivity patterns of

neurological and psychiatric patients but, also to investigate how local stimulation of

focal brain regions can alter those same connectivity patterns, in vivo. This method can

thus improve our overall understanding of brain functioning, while also refining where

and how neurostimulation treatments, such as TMS, are offered to patient populations.

Several reviews have assessed the capabilities of combining TMS with neuroimaging

tools (Bohning et al., 2003a; Bestmann et al., 2008a; Siebner et al., 2009; Vila-



https://doi.org/10.1101/2021.05.19.444738


5

Rodriguez et al., 2019; Bergmann et al., 2021), and discussed the implicated challenges

and limitations. Among others, some of the advantages is the possibility to assess

neuronal activity changes brought on by TMS using concurrent fMRI, mapping causal

top-down changes and interactions, and investigating the causal relationship between

brain region activations and behaviour (Bestmann et al., 2008a).

1.1.2. What do we know about artefact control

Combining the physics of TMS and fMRI brings on a number of technical challenges

that require consideration. One of these challenges is that the MRI scanner generates a

stable and homogenous magnetic field, while the TMS apparatus uses an unstable and

non-constant magnetic field. Therefore, in order to achieve proper concurrent TMS-

fMRI, specialized equipment (e.g., MR-compatible TMS machine, and customized TMS

and MR coils for concurrent TMS-fMRI) (Bohning et al., 2003b; Moisa et al., 2008;

Navarro De Lara et al., 2015; Cobos Sánchez et al., 2020) and synchronization

between MR sequences and TMS pulses (Bohning et al., 1999; Jung et al., 2016) is

required. Some of the other technical challenges include static and dynamic artifacts.

Static artefacts stem from the physical presence of the TMS apparatus in the MR

scanner (Bungert and Bowtell, 2008; Bungert et al., 2012), while dynamic artifacts arise

from using the TMS machine during and fMRI scan (e.g., radiofrequency, RF, noise;

leakage currents) (Weiskopf et al., 2009).

1.2. Objective

Given the technical complexities involved in combining TMS with neuroimaging

methods such as fMRI, it is important to investigate the dynamical artefacts that can be

generated by using TMS in an MR environment, prior to leading concurrent TMS-fMRI

experiments in human participants. Therefore, the objective of this study was to

measure the effect of TMS pulses, concurrently with fMRI sequencing, on the quality of

the fMRI images. Specifically, we examined the level of noise induced by the presence

of TMS and by TMS pulses on the acquired images of a phantom at different levels of



https://doi.org/10.1101/2021.05.19.444738


6

TMS pulse intensity. Although some studies have previously tackled this question at

their own institution levels (Moisa et al., 2008; Andoh and Zatorre, 2012), replication at

our own site, at the University of British Columbia (UBC), was necessary to ensure the

validity and reliability of the quality of images produced by concurrent TMS-fMRI, prior

to the start of clinical trials in patient populations.

2. Methods

2.1. Concurrent TMS-fMRI set up and stimulation protocol

This study was built on a standardized setup for concurrent TMS-fMRI at the UBC

MRI centre. Echo-planar images were collected from a BRAINO phantom on a Philips

Achieva 3.0 Tesla scanner. An array of two single channel SENSE Flex-L coils were

used in place of a volume head coil to allow optimal positioning of TMS coil. To reduce

any static artefact, a MR-compatible TMS coil was acquired (MRI-B91 non-

ferromagnetic TMS figure-8 coil), along with a high-current filter box that blocked current

leakage and filtered radio-frequency (RF) noise. In addition, as it has been suggested in

the literature, the distance between the trigger and the MRI scanner was maximized by

placing the TMS stimulator (MagPro X100) outside of the MR-room using an 8-meter

extension cable. The set-up of the phantom and the TMS coil is shown in Figure 1B.

To determine the optimal TMS pulse firing time, an adjustable delay period was

programmed into the synchronization hardware, which works as a relay between the

TMS coil pulses and the MRI sequences. Specifically, the synchronization box first

detects the single pulse generated by the MRI Philipps scanner at the beginning of each

functional sequence. Next, the box amplifies this pulse to relay it to the TMS stimulator

outside of the MRI room, in order to initiate a TMS pulse. During the concurrent TMS-

fMRI scan, each echo-planar volume was acquired in 2400 ms. Previous literature has

shown that if a period of at least 100 ms elapses after each TMS pulse, the subsequent

MR images are distortion free (Bohning et al., 1999). Accordingly, the rTMS pulse was

applied 2100 ms after the acquisition of the first slice, therefor allowing for an interval of

300 ms between the TMS pulse and the subsequent rs-MRI data acquisition (see Figure



https://doi.org/10.1101/2021.05.19.444738


7

1A). The synchronization between the TMS stimulator and echo planar image signal

was achieved using an in-house developed software and synchronization hardware

(partnership with PLC Electronics Solutions Ltd, Burnaby, Canada).

Figure 1. A) shows the stimulation protocol to test for potential artefacts when employing concurrent TMS and fMRI. B) illustrates the physical set-up of the phantom with the TMS coil inside the MRI scanner. EPI = echo planar imaging, fMRI = functional magnetic resonance imaging, TMS = transcranial magnetic stimulation, TR = repetition time

2.2. Stimulation protocol

The BRAINO spherical phantom was used to measure and compare the potential

artefacts generated by the use of concurrent TMS-fMRI on the functional images. To

evaluate image artefacts, the following conditions were examined: sequences (1)

without TMS coil attached; (2) with TMS coil attached (in the zenithal position in direct

contact with phantom), but with the stimulator turned off; (3) with TMS coil attached, and

7

al

d



https://doi.org/10.1101/2021.05.19.444738


8

the stimulator turned to 0% of the maximum TMS stimulator output (i.e., 0% intensity);

(4) with TMS coil attached, stimulating one pulse per time-repetition (i.e. one TMS pulse

per MRI volume), and the stimulator turned to 30% intensity; and (5) with TMS coil

attached, stimulating with one pulse per time-repetition, and the stimulator turned to

60% intensity. This sequential stimulation approach was taken to add one potential

artefact factor with each condition, in order to better isolate and quantify the potential

sources of image distortions. Specifically, condition (1) was the control condition, where

any artefact generated at this stage was attributable to the MRI scanner noise itself.

Condition (2) allowed to examine the potential artefact generated by the physical

presence of the TMS coil in the MR- room. Condition (3) allowed to investigate the

potential artefacts stemming from the TMS coil being energized. Finally, conditions (4)

and (5) allowed to evaluate the effect of two different intensities of TMS pulses on the

acquired images of the phantom.

2.3. Imaging Parameters and analyses

Each fMRI scans used a repetition time (TR) of 2400ms, a voxel size of 4mm × 4mm

× 4.5mm, for a total of 210 phantom volumes, as shown in Figure 1A. To assess

distortions on functional image qualities induced by using concurrent TMS-fMRI, several

in-house Matlab scripts were developed.

Among the measures used to evaluate the quality of the images, first, the signal image,

was computed by averaging the images obtained across the entire scanning session.

Second, the temporal fluctuation noise image was calculated as the standard deviation

of the residuals of the original, after subtracting the fit line (2nd order polynomial). Third,

the signal to fluctuation noise ratio (SFNR) was calculated by dividing the signal image

by the temporal fluctuation noise image. Finally, the spatial noise image was derived via

the difference of the averaged even-numbered fMRI volumes and the average of the

odd-numbered fMRI volumes. SFNR summary values were computed as the average

SFNR across voxels in the region of interest (ROI), for each condition, and was used to

quantify the temporal noise of the fMRI data. In the present report, the ROIs included



https://doi.org/10.1101/2021.05.19.444738


9

the entire phantom and its individual slices (see below). The analyses of the followed

those of a previously validated report (Friedman and Glover, 2006).

3. Results

The results of the analyses of the image distortions due to the use of concurrent

TMS-fMRI on a phantom are summarized in Figures 2 and 3. Figure 2 shows how the

whole phantom image (the mean image of the 210 volumes acquired on each

sequence), is altered across the five conditions in terms of the spatial noise and the

temporal fluctuation noise. Overall, the images of the phantom did not show any

obvious distortion under any of the five testing conditions, qualitatively. This held also

true for the temporal fluctuation noise and the spatial noise.

Figure 2. Signal and noise images for each test condition. A) shows the signal image and the arrow indicated the contact location between the phantom and the TMS coil. B) shows the spatial noise and C) shows the temporal fluctuation noise for each condition.

9



https://doi.org/10.1101/2021.05.19.444738


10

Across all conditions, the whole-phantom SFNR summary value remained fairly stable,

across the five testing conditions (Figure 3A). According to previous studies (Friedman

and Glover, 2006), SFNR summary values of a regular fMRI image are typically around

150-200 in a 3T MRI scanner. In our experiment, the SFNR values across the five

conditions ranged from 204.95 (condition 5) to 232.86 (condition 1), indicating that our

fMRI data were of reasonable quality even with TMS firing on a relatively high intensity.

Next, we examined more closely how the SFNR changed in each condition, as a

function to the proximity to the TMS coil. For Figure 3B, the ROI represents each 4mm

slice, with slice 1 being the slice closest to the TMS coil, up to slice number 20, where

the SFNR had stabilized. The results show that noise is maximal (low SFNR) at slice 1,

near the TMS coil, and that it improves and stabilises by slice 4. To make a parallel with

the human brain, slice 4 implies that we need to be approximately 16mm (slice 4, each

slice 4mm thick = 16mm) away from the coil to obtain good SFNR. It has been

estimated that the cortex lies approximately 10-20mm from the top of the scalp (Kozel et

al., 2000; Beauchamp et al., 2011). Moreover, examining the changes per slices, Figure

3 B also illustrate that conditions 2-5 overlap significantly in terms of the slice-wise

SFNR, and are slightly lower than those in the first condition, where the TMS apparatus

is absent from the MR-room.



https://doi.org/10.1101/2021.05.19.444738


11

Figure 3. SFNR summary values (each slice as ROI) for each condition. A) The phantom mask in the left panel reveals the contact point between the phantom and the TMS coil (arrow) Moreover, we can see the change in SNFR as the condition changes, and as more potential artefact is added (from no TMS stimulation to TMS simulation at 60% intensity). Error bars indicate the standard deviation of the SFNR of the entire phantom voxels. B) shows how the SFNR changes in each condition for the first 20 slices of the masked phantom. Error bars indicate the standard deviation of the SFNR of all voxels included in the slice. SFNR=signal to fluctuation noise ratio, TMS = transcranial magnetic stimulation

4. Discussion

This proof-of-concept phantom study demonstrated how concurrent TMS-fMRI

procedures were implemented at the Neuroimaging facility of the University of British

Columbia, for the purpose of using concurrent TMS-fMRI to better understand and treat

TRD. Specifically, the use of an MRI-compatible TMS apparatus, a high-current filter

box and physical separation of the TMS stimulator from the TMS coil helped reduce any

potential static or dynamic artefacts that could have risen from the combination of these

two neuroimaging tools (Ge et al., 2017). Given that the MRI scanner relies on a stable,

constant, magnetic field, while the TMS apparatus relies on unstable magnetic field, we

investigated where in the installation of the concurrent TMS-fMRI setup could potential

1

t

y

e

,



https://doi.org/10.1101/2021.05.19.444738


12

artefacts arise from. To do so, a phantom was used to investigate when the TMS was

not in the MRI room, when the TMS coil was in the MR-room but OFF, when the TMS

coil was on ON, but at 0% pulse intensity, and when the TMS coil was turned ON to

30% and 60% intensity, respectively. The analyses of these sequential conditions

revealed that minimal image distortion took place in conditions 2-5, implying that while

noise was slightly increased due to the presence of the TMS coil and cable in the MR-

room, the different intensities did not significantly affect the quality of the phantom

volumes. Importantly, we observed that the SFNR increased and stabilized when the

TMS coil was around 16mm from the site of interest, implying that temporal variability of

the EPI signal reduced as the distance with the coil increased. As one would find it, this

distance is approximately the distance between the human scalp and the cortex. This

finding suggests that robust, low noise, signal can be obtained using concurrent TMS-

fMRI at the cortical brain level. These findings replicate those other labs and

researchers (Moisa et al., 2008; Andoh and Zatorre, 2012).

Despite these great strides, it is important to keep in mind that in human participants,

TMS experiments can induce a startle or sensory response in the individual. These

effects, and changes in the intensity or site of the stimulation during fMRI sequencing

should be taken into consideration and investigated.

5. Conclusion

In sum, this study demonstrated the feasibility and artefact control involved in combining

the spatial resolution of fMRI with the interventional functions of TMS at the University of

British Columbia. This step was a critical step towards the clinical protocols aimed at

treating TRD.



https://doi.org/10.1101/2021.05.19.444738


13

REFERENCES

Andoh, J., and Zatorre, R. J. (2012). Mapping the After-effects of Theta Burst

Stimulation on the Human Auditory Cortex with Functional Imaging. J. Vis. Exp.

doi:10.3791/3985.

Beauchamp, M. S., Beurlot, M. R., Fava, E., Nath, A. R., Parikh, N. A., Saad, Z. S., et

al. (2011). The developmental trajectory of brain-scalp distance from birth through

childhood: Implications for functional neuroimaging. PLoS One 6.

doi:10.1371/journal.pone.0024981.

Bergmann, T. O., Varatheeswaran, R., Hanlon, C. A., Madsen, K. H., Thielscher, A.,

and Siebner, H. R. (2021). Concurrent TMS-fMRI for causal network perturbation

and proof of target engagement. Neuroimage 237, 118093.

doi:10.1016/j.neuroimage.2021.118093.

Bestmann, S., Oliviero, A., Voss, M., Dechent, P., Lopez-Dolado, E., Driver, J., et al.

(2006). Cortical correlates of TMS-induced phantom hand movements revealed

with concurrent TMS-fMRI. Neuropsychologia 44, 2959–2971. Available at:

http://eutils.ncbi.nlm.nih.gov/entrez/eutils/elink.fcgi?dbfrom=pubmed&id=16889805

&retmode=ref&cmd=prlinks.

Bestmann, S., Ruff, C. C., Blankenburg, F., Weiskopf, N., Driver, J., and Rothwell, J. C.

(2008a). Mapping causal interregional influences with concurrent TMS-fMRI. Exp.

Brain Res. 191, 383–402. doi:10.1007/s00221-008-1601-8.

Bestmann, S., Swayne, O., Blankenburg, F., Ruff, C. C., Haggard, P., Weiskopf, N., et

al. (2008b). Dorsal premotor cortex exerts state-dependent causal influences on

activity in contralateral primary motor and dorsal premotor cortex. Cereb. Cortex

18, 1281–91. doi:10.1093/cercor/bhm159.

Bohning, D. E., Denslow, S., Bohning, P. A., Lomarev, M. P., and George, M. S.

(2003a). “Chapter 5 Interleaving fMRI and rTMS,” in Transcranial Magnetic

Stimulation and Transcranial Direct Current Stimulation, eds. W. Paulus, F. Tergau,

M. A. Nitsche, J. G. Rothwell, U. Ziemann, and M. B. T.-S. to C. N. Hallett

(Elsevier), 42–54. doi:https://doi.org/10.1016/S1567-424X(09)70208-9.

Bohning, D. E., Denslow, S., Bohning, P. A., Walker, J. A., and George, M. S. (2003b).



https://doi.org/10.1101/2021.05.19.444738


14

A TMS coil positioning/holding system for MR image-guided TMS interleaved with

fMRI. Clin. Neurophysiol. 114, 2210–9. doi:10.1016/S1388-2457(03)00232-3.

Bohning, D. E., Pecheny, A. P., Epstein, C. M., Speer, A. M., Vincent, D. J., Dannels,

W., et al. (1997). Mapping transcranial magnetic stimulation (TMS) fields in vivo

with MRI. Neuroreport 8, 2535–2538. doi:10.1097/00001756-199707280-00023.

Bohning, D. E., Shastri, A., McConnell, K. A., Nahas, Z., Lorberbaum, J. P., Roberts, D.

R., et al. (1999). A combined TMS/fMRI study of intensity-dependent TMS over

motor cortex. Biol. Psychiatry 45, 385–394. doi:10.1016/S0006-3223(98)00368-0.

Bungert, a, and Bowtell, R. (2008). Effects of the TMS coil on MR image quality in

combined TMS/fMRI. Proc. 16th Sci. Meet. Int. Soc. Magn. Reson. Med. Toronto,

3028.

Bungert, A., Chambers, C. D., Phillips, M., and Evans, C. J. (2012). Reducing image

artefacts in concurrent TMS/fMRI by passive shimming. Neuroimage 59, 2167–

2174. doi:10.1016/j.neuroimage.2011.10.013.

Buxton, R. B. (2009). Introduction to Functional Magnetic Resonance Imaging:

Principles and Techniques. second edition. New York: Cambridige University Press

Available at:

https://books.google.ca/books?hl=en&lr=&id=vqkgAwAAQBAJ&oi=fnd&pg=PP8&dq

=functional+magnetic+resonance+imaging&ots=4rano7Q_jh&sig=3Ln-

EQp1my0rKZekhYSpG04VO3k&redir_esc=y#v=onepage&q=functional magnetic

resonance imaging&f=false [Accessed May 13, 2021].

Cobos Sánchez, C., Cabello, M. R., Olozábal, Á. Q., and Pantoja, M. F. (2020). Design

of TMS coils with reduced Lorentz forces: application to concurrent TMS-fMRI. J.

Neural Eng. 17, 016056. doi:10.1088/1741-2552/ab4ba2.

de Vries, P. M., de Jong, B. M., Bohning, D. E., Hinson, V. K., George, M. S., and

Leenders, K. L. (2012). Reduced parietal activation in cervical dystonia after

parietal TMS interleaved with fMRI. Clin. Neurol. Neurosurg. 114, 914–21.

doi:10.1016/j.clineuro.2012.02.006.

de Vries, P. M., de Jong, B. M., Bohning, D. E., Walker, J. A., George, M. S., and

Leenders, K. L. (2009). Changes in cerebral activations during movement

execution and imagery after parietal cortex TMS interleaved with 3T MRI. Brain



https://doi.org/10.1101/2021.05.19.444738


15

Res. 1285, 58–68. doi:10.1016/j.brainres.2009.06.006.

Denslow, S., Lomarev, M., George, M. S., and Bohning, D. E. (2005). Cortical and

subcortical brain effects of transcranial magnetic stimulation (TMS)-induced

movement: an interleaved TMS/functional magnetic resonance imaging study. Biol.

Psychiatry 57, 752–60. doi:10.1016/j.biopsych.2004.12.017.

Friedman, L., and Glover, G. H. (2006). Report on a multicenter fMRI quality assurance

protocol. J. Magn. Reson. Imaging 23, 827–839. doi:10.1002/jmri.20583.

Ge, R., Dipinto, A., Barlow, L., Macmillan, E., Mackay, A., Alfonso, M., et al. (2017). A

primer on developing interleaved transcranial magnetic stimulation with functional

magnetic resonance imaging capability: the UBC experience. Brain Stimul. Basic,

Transl. Clin. Res. Neuromodulation 10, 520–521. doi:10.1016/j.brs.2017.01.520.

George, M. S., Lisanby, S. H., and Sackeim, H. A. (1999). Transcranial magnetic

stimulation: Applications in neuropsychiatry. Arch. Gen. Psychiatry 56, 300–311.

doi:10.1001/archpsyc.56.4.300.

Hallett, M. (2000). Transcranial magnetic stimulation and the human brain. Nature 406,

147–150. doi:10.1038/35018000.

Hanlon, C. A., Canterberry, M., Taylor, J. J., DeVries, W., Li, X., Brown, T. R., et al.

(2013). Probing the frontostriatal loops involved in executive and limbic processing

via interleaved TMS and functional MRI at two prefrontal locations: a pilot study.

PLoS One 8, e67917. doi:10.1371/journal.pone.0067917.

Huettel, S. A., Song, A. W., and Mccarthy, G. (2004). Functional Magnetic Resonance

Imaging . second edition. Sunderland, MA: Sinauer Associates.

Jung, J., Bungert, A., Bowtell, R., and Jackson, S. R. (2016). Vertex Stimulation as a

Control Site for Transcranial Magnetic Stimulation: A Concurrent TMS/fMRI Study.

Brain Stimul. 9, 58–64. doi:10.1016/j.brs.2015.09.008.

Kozel, F. A., Nahas, Z., DeBrux, C., Molloy, M., Lorberbaum, J. P., Bohning, D., et al.

(2000). How coil-cortex distance relates to age, motor threshold, and

antidepressant response to repetitive transcranial magnetic stimulation. J.

Neuropsychiatry Clin. Neurosci. 12, 376–384. doi:10.1176/jnp.12.3.376.

Li, X., Large, C. H., Ricci, R., Taylor, J. J., Nahas, Z., Bohning, D. E., et al. (2011).

Using interleaved transcranial magnetic stimulation/functional magnetic resonance



https://doi.org/10.1101/2021.05.19.444738


16

imaging (fMRI) and dynamic causal modeling to understand the discrete circuit

specific changes of medications: Lamotrigine and valproic acid changes in motor or

prefrontal e. Psychiatry Res. - Neuroimaging 194, 141–148.

doi:10.1016/j.pscychresns.2011.04.012.

Li, X., Tenebäck, C. C., Nahas, Z., Kozel, F. A., Large, C., Cohn, J., et al. (2004).

Interleaved transcranial magnetic stimulation/functional MRI confirms that

lamotrigine inhibits cortical excitability in healthy young men.

Neuropsychopharmacology 29, 1395–1407. Available at:

http://eutils.ncbi.nlm.nih.gov/entrez/eutils/elink.fcgi?dbfrom=pubmed&id=15100699

&retmode=ref&cmd=prlinks.

Logothetis, N. K. (2008). What we can do and what we cannot do with fMRI. Nature

453, 869–878. doi:10.1038/nature06976.

Moisa, M., Pohmann, R., Ewald, L., and Thielscher, A. (2009). New coil positioning

method for interleaved transcranial magnetic stimulation (TMS)/functional MRI

(fMRI) and its validation in a motor cortex study. J. Magn. Reson. Imaging 29, 189–

197. doi:10.1002/jmri.21611.

Moisa, M., Pohmann, R., Uluda�, K., and Thielscher, A. (2010). Interleaved TMS/CASL:

Comparison of different rTMS protocols. Neuroimage 49, 612–620.

doi:10.1016/j.neuroimage.2009.07.010.

Moisa, M., Uludag, K., Ugurbil, K., and Thielscher, A. (2008). A new hardware-software

coil positioning system for interleaved TMS/fMRI: A motor cortex stimulation study.

Proc. 16th Sci. Meet. Int. Soc. Magn. Reson. Med. Toronto, 2428. Available at:

/MyPathway2008/2428.

Nahas, Z., Lomarev, M., Roberts, D. R., Shastri, A., Lorberbaum, J. P., Teneback, C., et

al. (2001). Unilateral left prefrontal transcranial magnetic stimulation (TMS)

produces intensity-dependent bilateral effects as measured by interleaved BOLD

fMRI. Biol. Psychiatry 50, 712–20. doi:10.1016/S0006-3223(01)01199-4.

Navarro De Lara, L. I., Windischberger, C., Kuehne, A., Woletz, M., Sieg, J., Bestmann,

S., et al. (2015). A novel coil array for combined TMS/fMRI experiments at 3 T.

Magn. Reson. Med. 74, 1492–1501. doi:10.1002/mrm.25535.

Oathes, D. J., Zimmerman, J. P., Duprat, R., Japp, S. S., Scully, M., Rosenberg, B. M.,



https://doi.org/10.1101/2021.05.19.444738


17

et al. (2021). Resting fMRI-guided TMS results in subcortical and brain network

modulation indexed by interleaved TMS/fMRI. Exp. brain Res. doi:10.1007/s00221-

021-06036-5.

Ricci, R., Salatino, A., Li, X., Funk, A. P., Logan, S. L., Mu, Q., et al. (2012). Imaging the

neural mechanisms of TMS neglect-like bias in healthy volunteers with the

interleaved TMS/fMRI technique: preliminary evidence. Front. Hum. Neurosci. 6,

326. doi:10.3389/fnhum.2012.00326.

Schintu, S., Cunningham, C. A., Freedberg, M., Taylor, P., Gotts, S. J., Shomstein, S.,

et al. (2021). Callosal anisotropy predicts attentional network changes after parietal

inhibitory stimulation. Neuroimage 226, 117559.

doi:10.1016/j.neuroimage.2020.117559.

Siebner, H. R., Bergmann, T. O., Bestmann, S., Massimini, M., Johansen-Berg, H.,

Mochizuki, H., et al. (2009). Consensus paper: combining transcranial stimulation

with neuroimaging. Brain Stimul. 2, 58–80. doi:10.1016/j.brs.2008.11.002.

Vila-Rodriguez, F., Downar, J., and Blumberger, D. M. (2013). Repetitive Transcranial

Magnetic Stimulation in Depression: A Changing Landscape | Depression [1],

Transcranial Magnetic Stimulation [2]. Available at: http://www.psychiatrictimes.com

[Accessed May 13, 2021].

Vila-Rodriguez, F., Ge, R., and Long, D. (2019). Interleaved Transcranial Magnetic

Stimulation and Functional Magnetic Resonance Imaging: A Translational Tool.

Clin. Pharmacol. Ther. 106, 714–716. doi:10.1002/cpt.1575.

Webler, R. D., Hamady, C., Molnar, C., Johnson, K., Bonilha, L., Anderson, B. S., et al.

(2020). Decreased interhemispheric connectivity and increased cortical excitability

in unmedicated schizophrenia: A prefrontal interleaved TMS fMRI study. Brain

Stimul. 13, 1467–1475. doi:10.1016/j.brs.2020.06.017.

Weiskopf, N., Josephs, O., Ruff, C. C., Blankenburg, F., Featherstone, E., Thomas, A.,

et al. (2009). Image artifacts in concurrent transcranial magnetic stimulation (TMS)

and fMRI caused by leakage currents: modeling and compensation. J. Magn.

Reson. Imaging 29, 1211–7. doi:10.1002/jmri.21749.



https://doi.org/10.1101/2021.05.19.444738


implementation of a tms-fmri system: a primer · 2021. 5. 19. · navarro de lara et al., 2015;...

Documents